ResearchBenchmarksQuestion Answering on JeopardyFollow34.7AccuracyEnsemble26.06828.30930.5532.791Jun 17, 2025Evaluation ResultsMethodMethodLinksAccuracyEnsembleAggregation Method=Byt...Aggregation Method=Byte-level Ensemble2025.0634.7OLMO2Model=OLMO2Model=OLMO22025.0632.7AverageAggregation Method=Ave...Aggregation Method=Average2025.0628.8QWEN3Model=QWEN3Model=QWEN32025.0627.4LLAMA3.2Model=LLAMA3.2Model=LLAMA3.22025.0626.4